There are two ways in which memory is most likely stored. The first is
simply a correlation of activity between neurons. This can be anything from
an oscillating circuit to simply synchronizing the firing of many neurons.
The amount of time that information can be stored in this manner is a matter
of debate among neuroscientists. However, mathematicians (see Langton
Physica D 42:12-37, 1990) claim that the length of time that information can
be stored in this manner is dependant on the size of the network and,
theoretically, the storage could be infinite, or at least as long as one's
life time.
The second mechanism is much more heavily studied and involves storing
information within the synapses by adjusting their strength. This is
thought to occur by increasing the amount of transmitter released from the
presynaptic terminal or by increasing the postsynaptic response to a given
amount of transmitter, or through both mechanisms (see Churchland and
Sejnowski, The Computational Brain). This is the mechanism of information
storage implied when people study the phenomena of long term potentiation
and long term depression. This form of information storage has already been
exploited for numerous applications in the use of neural nets.
There may also be other forms of information storage in the brain that are
currently poorly understood or have not even been conceived of yet, thus the
field of information storage in nervous systems should remain an important
research topic for many years.
R. Caudle
